Multi-Agent Collaboration System Skills
PassAudited by VirusTotal on May 14, 2026.
Findings (1)
```json { "classification": "suspicious", "summary": "The skill bundle is classified as suspicious due to a shell injection vulnerability in `scripts/init.sh`. The `sed` command used to replace the project name placeholder (`s/\[项目名称\]/$PROJECT_NAME/g`) directly interpolates user-provided input (`$PROJECT_NAME`) without proper escaping. This could allow an attacker to inject arbitrary shell commands or manipulate the `llms.txt` file if the `project-name` argument contains special characters like `/` or newlines. While the overall purpose of the skill (multi-agent collaboration and document management) appears benign, this vulnerability presents a significant security risk." } ```
